New England Journal of Medicine October 24, 2026

Mechanical Thrombectomy 24 to 72 Hours after Large-Vessel Ischemic Stroke

Pengfei Yang, Hongjian Shen, Tingyu Yi, Pengfei Xing, Junfeng Su, et al.

Bottom Line

In imaging-selected patients with large-vessel ischemic stroke presenting 24 to 72 hours after onset, mechanical thrombectomy plus medical care improved functional outcomes at 90 days compared with medical care alone, but it increased the risk of symptomatic intracranial hemorrhage and early mortality.

Key Findings

1. Endovascular therapy was associated with a favorable shift in the distribution of functional outcomes on the modified Rankin scale at 90 days (common odds ratio for better outcome, 1.74; 95% CI, 1.19 to 2.55; P=0.005).

2. Death or dependency at 90 days occurred in 51.5% of the thrombectomy group versus 64.7% of the medical management group (odds ratio, 0.58).

3. Symptomatic intracranial hemorrhage was significantly more common in the intervention group compared with the control group (4.8% vs. 0%).

4. Mortality by 90 days was increased in the thrombectomy group compared with the control group (6.0% vs. 1.8%, odds ratio 3.48).

5. Procedural complications in the intervention group included dissection or perforation in 4.8% of patients and embolization to a new territory in 2.4%.

Study Design

Design

RCT

Open-Label

Sample

336

Patients

Duration

90 days

Median

Setting

Multicenter, China

Population Adults with anterior-circulation large-vessel occlusion ischemic stroke presenting 24 to 72 hours after symptom onset, with salvageable brain tissue confirmed by automated perfusion imaging (defined as an ischemic core <50 mL, mismatch ratio ≥1.8, and penumbra ≥15 mL).

Intervention Mechanical thrombectomy plus standard medical management.

Comparator Standard medical management alone.

Outcome Functional outcome at 90 days, assessed by the ordinal shift in scores on the modified Rankin scale.

Study Limitations

• The open-label design may have introduced performance and management bias, as evidenced by the intervention group receiving substantially more intravenous tirofiban, osmotic therapy, and a 2.4-day longer hospitalization.

• The trial was conducted exclusively in Chinese comprehensive stroke centers, raising generalizability concerns regarding broader application to North American and European populations.

• The clear benefits in functional independence came with a clinically meaningful increased risk of symptomatic intracranial hemorrhage and early mortality, demanding rigorous imaging-based patient selection.

Clinical Significance

The LATE-MT trial establishes that strict perfusion imaging criteria can identify patients with salvageable brain tissue who benefit from mechanical thrombectomy up to 72 hours after stroke onset. This extends the treatment window far beyond the standard 24 hours, representing a massive paradigm shift from a time-based to a tissue-based approach, although clinicians must carefully weigh this functional benefit against an elevated risk of severe hemorrhagic complications and death.

Historical Context

Following landmark trials in 2015 that standardized thrombectomy within 6 hours of stroke onset, the DAWN and DEFUSE 3 trials (2018) revolutionized acute stroke care by using perfusion imaging to extend the window up to 24 hours. LATE-MT pushes this boundary even further to 72 hours, addressing a major evidence gap for late-presenting patients and affirming that physiological criteria (ischemic penumbra) rather than a rigid chronological clock should dictate intervention eligibility.

Guided Discussion

High-yield insights from every perspective

Med Student

Medical Student

What is the pathophysiological concept of the ischemic penumbra, and how do advanced imaging modalities like CT perfusion identify patients who might still benefit from thrombectomy up to 72 hours after symptom onset?

Key Response

Understanding the mismatch between cerebral blood flow (representing the unsalvageable infarct core) and delayed transit time (representing the salvageable penumbra) is crucial. Collateral circulation can preserve this penumbra for days in 'slow progressors', making late intervention possible if accurately identified on imaging.

Resident

Given the increased risk of symptomatic intracranial hemorrhage (sICH) and early mortality in the thrombectomy group, how should this influence your informed consent discussion and immediate post-procedural management for a patient presenting at 48 hours with a large-vessel occlusion?

Key Response

Residents must balance the potential for long-term functional independence against acute risks of sICH. Counseling families requires translating these dual findings into shared decision-making, while post-procedural management requires extremely strict hemodynamic control to mitigate reperfusion injury.

Fellow

How might variations in baseline leptomeningeal collateral status and infarct core growth rates explain the juxtaposition of improved 90-day functional outcomes alongside increased early mortality and sICH in this late-window population?

Key Response

Fellows must grasp that 'slow progressors' with robust collaterals remain eligible at 72 hours, but late reperfusion into vascular beds that have been ischemic for days significantly increases the risk of blood-brain barrier breakdown and hemorrhagic transformation (reperfusion injury).

Attending

If extending the thrombectomy window to 72 hours becomes a standard option for imaging-selected patients, how will this impact regional stroke systems of care, and what specific institutional protocols must be implemented to minimize the observed early mortality?

Key Response

Attendings must consider systems-level impacts. A 72-hour window dramatically increases the volume of patients eligible for 'drip and ship' advanced imaging transfers, straining resources. Mitigating mortality requires protocolized neurocritical care, stringent blood pressure parameters, and meticulous patient selection.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD

When evaluating functional outcomes via an ordinal modified Rankin Scale (mRS) shift at 90 days in a trial with significantly increased early mortality in the intervention arm, what statistical frameworks (such as the win ratio or specific estimands) are necessary to handle mortality as a competing risk and avoid survivorship bias?

Key Response

A critical methodological challenge is that if early mortality is higher in the treatment arm, standard mRS shift analyses can be skewed. Researchers must ensure that the statistical model appropriately penalizes death (mRS 6) so the functional benefit among survivors is not falsely exaggerated.

Journal Editor

What threats to external validity are introduced by the proprietary advanced imaging software and specific mismatch thresholds required for patient selection, and how robust is the blinding of the 90-day mRS assessors given the notable difference in sICH rates between groups?

Key Response

Editors should flag that reliance on highly specific AI-driven perfusion software available mostly at academic centers limits generalizability. Furthermore, the higher rate of acute complications like sICH in the treatment arm can easily inadvertently unblind mRS assessors, introducing detection bias.

Guideline Committee

How should AHA/ASA stroke guidelines incorporate this 24-72 hour window considering the trade-off between improved long-term disability and increased early mortality, and how does this evidence modify the existing Class I recommendations for the 6-24 hour window established by DAWN and DEFUSE-3?

Key Response

Current guidelines strongly recommend MT up to 24 hours based on DAWN/DEFUSE-3 criteria. Pushing the window to 72 hours with a known safety trade-off would likely result in a weaker Class IIa or IIb recommendation, mandating strict adherence to the trial's exact imaging parameters rather than widespread adoption.

Clinical Landscape

Noteworthy Related Trials

2015

MR CLEAN Trial

n = 500 · NEJM

Tested

Intraarterial treatment (predominantly mechanical thrombectomy)

Population

Patients with acute ischemic stroke due to proximal arterial occlusion within 6 hours of onset

Comparator

Usual care alone

Endpoint

Modified Rankin scale score at 90 days

Key result: Mechanical thrombectomy within 6 hours of symptom onset significantly improved the rate of functional independence.

2018

DAWN Trial

n = 206 · NEJM

Tested

Mechanical thrombectomy plus standard care

Population

Ischemic stroke patients with large-vessel occlusion 6-24 hours after last known well with clinical-core mismatch

Comparator

Standard medical care alone

Endpoint

Utility-weighted modified Rankin scale score at 90 days

Key result: Thrombectomy significantly improved functional outcomes at 90 days compared to medical care alone in patients with clinical-core mismatch.

2018

DEFUSE 3 Trial

n = 182 · NEJM

Tested

Endovascular thrombectomy plus standard medical therapy

Population

Ischemic stroke patients with large-vessel occlusion 6-16 hours after last known well with salvageable tissue on perfusion imaging

Comparator

Standard medical therapy alone

Endpoint

Modified Rankin scale score at 90 days

Key result: Endovascular therapy resulted in significantly better functional outcomes and lower mortality at 90 days than medical therapy alone.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis